feat(training): measured BP/vapor pressure from PubChem (public domain) - #56
Merged
Merged
Conversation
… PubChem Enables the "measured volatility" lookup with a commercial-clean source. PubChem experimental properties are public domain, so build_properties.py pulls each molecule's experimental Boiling Point (+ best-effort Vapor Pressure) via PUG-View, parses to °C / Pa (converting °F, averaging ranges, skipping measurement-condition temperatures), and writes properties.parquet — the table predict.py already reads. We use a MEASURED lookup on purpose: structure-based BP (Joback) was rejected as too inaccurate, so a number is only reported when it's a real measurement. Verified on live PubChem against known boiling points: vanillin 285 °C, eugenol 254 °C, menthol 215 °C, limonene 178 °C, ethyl acetate 77 °C — all correct. The data itself is gitignored (built locally); only the fetcher is committed. Signed-off-by: Austin L. <86896075+rvnminers-A-and-N@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
First of the two clean data wins from the completion plan — built and tested, not just documented.
PubChem experimental properties are public domain (no commercial restriction), so
training/build_properties.pypulls each molecule's experimental Boiling Point (+ best-effort Vapor Pressure) via PUG-View, parses to °C / Pa, and writesproperties.parquet— the lookup tablepredict.pyalready reads for measured volatility. (Structure-based BP was rejected as too inaccurate, so we only ever report a measured number.)Parser handles the real-world mess: °F→°C conversion, range averaging, and skipping measurement-condition temperatures (e.g. '…at 20 °C') that would otherwise be mistaken for the BP.
Verified live against known boiling points: vanillin 285 °C, eugenol 254 °C, menthol 215 °C, limonene 178 °C, ethyl acetate 77 °C — all correct.
ruffclean. Data is gitignored (built locally); only the fetcher is committed.Next clean win (documented, not yet built): GRAS cross-check from FDA SAF (public domain).